Google Professional Data Engineer Question 152
Select 2Google Cloud PlatformYou are designing a data architecture for a retail analytics platform on Google Cloud. The platform needs to support the following use cases:
- Real-time inventory updates for store managers (low-latency writes).
- Daily sales trend analysis for business analysts (high-throughput reads).
- Customer data retrieval to personalize shopping experiences (low-latency reads).
Which combination of Google Cloud services would best support these access patterns?
- A
Use BigQuery for daily sales trend analysis, Cloud Spanner for real-time inventory updates, and Firestore for customer data retrieval.
- B
Use Cloud SQL for real-time inventory updates, Bigtable for daily sales trend analysis, and Memorystore for customer data retrieval.
- C
Use Firestore for real-time inventory updates, BigQuery for daily sales trend analysis, and Memorystore for customer data retrieval.
- D
Use Bigtable for real-time inventory updates, BigQuery for daily sales trend analysis, and Cloud Spanner for customer data retrieval.
Show answer and explanation
Correct answers: A, C
Explanation
The correct architecture should use services optimized for the specific access patterns. BigQuery is well-suited for analytical workloads with high-throughput read requirements. Cloud Spanner and Firestore are both viable choices for real-time inventory updates, with Firestore offering a simpler NoSQL solution for structured data. Memorystore is ideal for customer data retrieval due to its low-latency caching capabilities. Selecting the appropriate services ensures efficient data access and processing across all use cases.
- A. Correct.
Correct: BigQuery is optimized for high-throughput analytical queries, Cloud Spanner handles globally-distributed low-latency writes, and Firestore is a good choice for low-latency reads and structured data retrieval.
- B. Incorrect.
Incorrect: Cloud SQL is not the best option for real-time inventory updates due to its lack of scalability for high-frequency writes, and Bigtable is not optimized for analytical queries.
- C. Correct.
Correct: Firestore works well for real-time inventory updates with its NoSQL capabilities, BigQuery is suitable for daily sales trend analysis, and Memorystore is ideal for low-latency reads of cached data like customer profiles.
- D. Incorrect.
Incorrect: Bigtable is not optimal for real-time inventory updates, and Cloud Spanner is not the best choice for low-latency reads as it is designed for transactional workloads.